DocumentCode
3209790
Title
Evolutionary design of ANN structure using genetic algorithm
Author
Moharamzade, N. ; Farokhi, F.
Author_Institution
Sci. Assoc. of Electr. & Electron. Eng., Islamic Azad Univ., Tehran, Iran
Volume
2
fYear
2010
fDate
13-14 Sept. 2010
Firstpage
219
Lastpage
224
Abstract
Determining the optimum structure for an Artificial Network is an important design step in almost all the artificial intelligence systems which are based on Neural or neuro-fuzzy networks. In this paper a genetic algorithm based solution is presented and tested over real world databases and for single layer and multiple layer networks and it has been shown that the determined network structures has the best accuracy and the optimized topology as well.
Keywords
fuzzy neural nets; genetic algorithms; multilayer perceptrons; artificial intelligence system; evolutionary ANN structure design; genetic algorithm; multiple layer network; neuro-fuzzy network; single layer network; Computational intelligence;
fLanguage
English
Publisher
ieee
Conference_Titel
Computational Intelligence and Natural Computing Proceedings (CINC), 2010 Second International Conference on
Conference_Location
Wuhan
Print_ISBN
978-1-4244-7705-0
Type
conf
DOI
10.1109/CINC.2010.5643748
Filename
5643748
Link To Document